Breast-conserving surgery: the balance between good cosmesis and local control

Introduction

The aim of local treatment of breast cancer is to achieve long-term local disease control with the minimum of local morbidity. The majority of women presenting symptomatically to breast clinics and those who are diagnosed through screening programmes have small breast cancers, which are suitable for breast-conserving therapy (BCT), defined as breast-conserving surgery (BCS) and whole-breast radiotherapy.

The major advantages of breast-conserving treatment are:

equivalence in terms of disease outcome compared with mastectomy as demonstrated in two systematic reviews ^, ;
an acceptable cosmetic appearance for the majority of women with breast cancer ;
fewer complications and more cost-effective than mastectomy; and
lower levels of psychological morbidity compared with mastectomy, with less anxiety and depression and improved body image, sexuality and self-esteem. ^,

The Early Breast Cancer Trialists Collaborative Group (search date 1995) analysed data from six randomised controlled trials (RCTs) that compared BCT with mastectomy. A meta-analysis of data from five of these six trials involving 3006 women found no significant difference in the risk of death at 10 years (odds ratio [OR] 0.91, 95% confidence interval [CI] 0.78–1.05). The sixth randomised trial used different protocols. A second systematic review included nine RCTs involving 4981 women randomised to mastectomy or BCT. A meta-analysis of these nine trials found no significant difference in the risk of death over 10 years: the relative risk (RR) reduction for BCT compared with mastectomy was 0.02 (95% CI − 0.05 to + 0.09). There was also no difference in the rates of local recurrence in the six RCTs involving 3006 women where data were available: the RR reduction for mastectomy versus BCT was 0.04 (95% CI − 0.04 to + 0.12). Longer-term follow-up of these trials did show an excess of local recurrences with BCT in four of the trials (OR 1.561, 95% CI 1.289–1.890, P < 0.001), but the pooled analysis for mortality showed no effect (OR 1.070, 95% CI 0.935–1.224, P = 0.33). Many of the later local events in treated patients are second breast cancers rather than true local recurrences and this may explain why even with the increase in local events there was no survival benefit for mastectomy. The current use of long-term adjuvant hormone therapy now prevents many of these new events.

These randomised trials comparing BCT with mastectomy were performed many years ago. Over the time period since these trials enrolled patients, recurrence rates have fallen dramatically. One reason for the continued high and increasing mastectomy rate in some countries is that patients and doctors continue to make decisions based on results from these older studies. Other potential reasons for a high rate of mastectomy include the increasing use of magnetic resonance imaging (MRI), ^, and in the United States, that has one of the highest rates of mastectomy, patient anxiety at leaving breast tissue, improvements in reconstructive techniques and the wish to avoid regular mammography. It should be considered essential to have an informed discussion with patients about the pros and cons of the various approaches if they are suitable for breast-conservation surgery but want mastectomy so patients can make a truly informed decision. Evidence from large cohort studies from many countries show that BCT survival and local recurrence rates are at least equivalent to and in many studies better than mastectomy. This applies to all cancer types including triple-negative breast cancers. Some of the studies that have reported better outcomes with BCT are detailed in Table 7.1 .

Table 7.1

Studies comparing BCT with mastectomy demonstrating better overall survival outcomes with BCT

Author	Country	No. of patients	RR	95% CI
Hwang et al.	USA	112 154	0.81	0.80–0.83
Lagendijk et al.	the Netherlands	129 692	0.74	0.71–0.76
Van Maaren et al.	the Netherlands	37 207	0.81	0.78–0.85
Hofvind et al.	Norway	9547	0.59	0.42–0.77
Hartmann-Johnsen et al.	Norway	13 015	0.61	0.53–0.70
Agarwal et al.	USA	132 149	0.76	0.72–0.78
Corradini et al.	Germany	7565 Cohort 1802	0.69	0.53–0.89
Onitilo et al.	Australia	5335	0.63	0.53–0.74
Chen et al.	USA	11 514 TNBC	0.58	0.49–0.69
Fisher et al.	Canada	14 633 stage II, III	0.74–0.57	0.41–0.88

BCT , Breast-conserving surgery; 95% CI , 95% confidence interval; RR , relative risk; TNBC , triple negative breast cancer.

The criticism of these studies is that in the main they are observational and some of the difference in outcomes is likely to be due to selection bias. The consistency of the findings and the findings that the better outcomes for BCT are consistent in screen-detected, interval and symptomatic cohorts is impressive. The rate of surgical complications and economic burden particularly with brachytherapy is better for BCT. Mastectomy has twice the rate of complications compared with BCT and is a much less cost-effective option than BCT, particularly when mastectomy is combined with breast reconstruction.

It seems unlikely that having mastectomy in itself is detrimental; however, the majority of patients undergoing BCT receive radiotherapy to the breast area, whereas only a minority of those having mastectomy do. It is therefore possible that the radiotherapy contributes to the apparent improvement in outcomes now seen with BCT compared with mastectomy alone. It could also be that the greater extent of surgery in patients having mastectomy is important. BCT does also offer the possibility of limiting the extent of axillary surgery for those with limited nodal involvement at sentinel node biopsy under criteria for the ACOSOG Z011 trial.

Originally it was thought that local therapy had little influence on overall survival, but it is clear that a proportion of local failures are responsible, at least in part, for some patients developing metastatic disease. ^,

It is thus important in patients selected for BCS to minimise local recurrence while at the same time achieving a good cosmetic outcome.

Selection of patients for breast conservation

Traditionally, single cancers measuring 4 cm or less, without signs of local advancement, have been selected for BCT. Currently any cancer, single or multiple, that can be excised to clear margins before or after neoadjuvant therapy or oncoplastic techniques alone or in combination and leaves a satisfactory cosmetic outcome can be treated with BCT ( Box 7.1 ).

Increasing tumour size does not equate with increasing local recurrence rate and so limiting BCS to cancers below a certain size is illogical.

Box 7.1

Indications and contraindications for breast-conserving surgery

Indications

Single or multiple lesions that can be excised to leave a satisfactory cosmetic outcome
Most T1, T2 (< 4 cm) cancers or T2 > 4 cm ∗

∗ Consider neoadjuvant therapy first.
T3 cancers in larger breasts ∗

Relative contraindications ^†

‡ Following a fully informed discussion of the pros and cons of breast-conserving surgery vs mastectomy.

† None of these are absolute contraindications.

T4, N2 or M1
Patients who prefer mastectomy ^‡
Collagen vascular disease ^§

§ Many patients with collagen vascular disease are suitable for wide excision and whole-breast radiotherapy.
Large or central tumours in small breasts ^¶

¶ Can be suitable for breast-conserving surgery after neoadjuvant treatment.
Women with a strong family history of breast cancer or who are proven BRCA1 and BRCA2 mutation carriers

Clinical measurements overestimate tumour size so this is best estimated by imaging, with ultrasound assessment of tumour size being more accurate than mammographic measurements. Ultrasound tends to underestimate pathologic size. MRI is better than ultrasound in assessing disease extent, particularly in invasive lobular carcinoma, but both over- and underestimation are seen even with MRI with measurement errors within a range of clinical importance in terms of their implications for the choice of treatment. MRI has a low specificity and a low positive predictive value, with only two-thirds of lesions identified by MRI as suspicious of malignancy subsequently confirmed as malignant. The role of MRI in assessing patients for BCS was investigated in a randomised study that showed that routine use of MRI did not produce any benefit. ^, In particular MRI did not reduce the rate of incomplete excision and was not associated with a reduction in short-term local recurrence but did significantly increase the mastectomy rate in patients randomised to MRI who were otherwise considered good candidates for BCS.

It is the balance between tumour size as assessed by imaging and breast volume that determines whether a patient is suitable for BCS.

Options for patients with tumours considered too large, relative to the size of the breast, for breast-conserving treatment include:

neoadjuvant systemic therapy to shrink the tumour,
an oncoplastic procedure (transfer of tissue into the breast or remodelling one or both breasts),
a combination of neoadjuvant treatment and oncoplastic surgery techniques to obtain symmetry, see Chapter 8 ). ^,

In a patient with small breasts, excision of even a small tumour may produce an unacceptable cosmetic result without tissue transfer.

Multisite Breast Cancers

Patients with multiple tumours in the same breast have not previously been considered good candidates for breast-conserving treatment because early studies reported a high incidence of in-breast recurrence, ^, and so patients with multiple tumours have traditionally been treated by mastectomy. The current nomenclature of using multifocal and multicentric is not useful for a number of reasons. Some multifocal cancers are further apart than some multicentric cancers. Almost two-thirds of multifocal and multicentric cancers are clonal and have the same genetic and genomic signature, indicating they are parts of the same cancer. Many multifocal cancers are unifocal, but the links between the different components parts of the cancer are not appreciated by either imaging or two-dimensional breast pathology. For this reason, we prefer the term ‘multisite cancers.’ A systematic review assessed the impact of BCS on the cancer outcomes of patients with multiple ipsilateral breast cancers. Comparative studies showed that local recurrence fell from the early series in patients with multiple cancers to levels now seen with unifocal cancers. Six studies compared BCS and mastectomy and showed equivalent rates of locoregional recurrence (LRR) (risk ratio 0.94, 95% CI 0.65 to 1.6). 64 (interquartile range [IQR] 57–73) months. One high-quality study reported 10-year actuarial LRR rates of 5.5% for BCS in 300 women versus 6.5% for mastectomy among 887 women. In view of these findings, there now appear to be similar rates of local recurrence for patients with unifocal and multisite disease providing all disease is excised to clear margins. If it is feasible to excise separate cancers in different parts of the breast and produce an acceptable cosmetic outcome then such patients should no longer be treated routinely by mastectomy. Breast conservation did prove feasible in the majority of patients with multisite cancer separated by 2 cm or greater in a recent US study. Some 67.6% of patients achieved a margin-negative excision in a single operation and only 7.1% of patients required conversion to mastectomy due to positive margins. Patients with bilateral cancers can also be treated by bilateral BCS.

The rates of BCS vary significantly within and between countries. These rates are clearly influenced as much by the views of the surgeon as by other issues such as the availability of radiotherapy locally. If a patient who fulfils the criteria for BCS is treated by mastectomy then the reasons for the decision to proceed to mastectomy should be recorded clearly in the patient’s notes. Some patients choose mastectomy in preference to BCS but may do so because they do not appreciate that outcomes for BCT are at least as good as mastectomy in both survival and recurrence rates. In one series of patients choosing mastectomy rather than BCS, over half of patients did not know that mastectomy and BCT produce identical rates of survival.

A range of clinical and pathological factors have influenced selection of patients for BCS because of their perceived impact on local recurrence. These include young age (under 35–39 years), the presence of an extensive in situ component (EIC) associated with an invasive tumour, grade 3 histology and widespread lymphatic/vascular invasion (LVI). These are considered in detail later.

Factors affecting local recurrence after BCS

Over 80% of all local recurrences were initially reported to be located adjacent to the site of initial excision. This is no longer true and an increasing percentage of ‘recurrences’ in treated breasts are second primary cancers.

Radiation therapy delivered to the whole breast continues to be used in most patients after BCS because radiotherapy both reduces the rate of local recurrence and improves overall survival. There is low-quality evidence from five RCTs ( N = 3407) that there is no effect of a difference in local recurrence-free survival at 5 to 10 years follow-up for women with invasive breast cancer when comparing partial breast radiotherapy with whole breast radiotherapy. The methods for delivering partial breast radiotherapy remains a topic of ongoing debate and are discussed in detail in Chapter 17 . It has not been possible to identify groups of patients who do not benefit from radiotherapy. However, there is a group of older patients with low-risk cancers (completely excised, node-negative and hormone receptor-rich on adjuvant hormone therapy) and women of any age whose cancers have an extremely good prognosis (small grade 1 or special-type cancers that are completely excised, node-negative and hormone receptor-positive on adjuvant hormone therapy) for whom the absolute risk of local recurrence at 5 years is very low. The benefits of giving radiotherapy to women with a very low absolute risk of recurrence remain uncertain if they are willing to take endocrine therapy. Following whole-breast radiotherapy, boosting the tumour bed reduces local recurrence rates, particularly in younger women and women with ductal carcinoma in situ (DCIS) out with the invasive cancer, although there are cosmetic penalties associated with the use of boost. Further discussion on these issues can be found in Chapter 17 .

The rates of in-breast tumour recurrences (IBTRs) following BCT have reduced dramatically over the past two decades ^, ( Fig. 7.1 ). Whereas a 1% annual rate of in-breast cancer events was formerly considered acceptable, rates are often now less than 0.25% per annum. IBTR rates for women with BCT do, however, remain at this rate for at least 20 years after treatment. This needs to be borne in mind when considering surveillance programs for such patients. Even patients treated by mastectomy are at risk of local recurrence over this 20-year period.

Figure 7.1, Local recurrence rates in Edinburgh over four separate time periods showing a significant and continued fall in local recurrence rates over time.

Patient-related factors

Recent studies do not support avoidance of BCT in younger women in whom clear margins can be achieved.

Multiple studies have demonstrated that local recurrence after BCS is more common in younger women. However, a recent meta-analysis of studies of women under age 40 undergoing BCT or mastectomy for breast cancer demonstrated no difference in overall survival (HR 0.90, 95% CI 0.81–1.00), with actually a trend in favour of BCT. A further review demonstrated that the recurrence-free survival for mastectomy and BCT was identical. These observations do not support avoidance of BCT in younger patients in whom clear margins can be achieved, and while patients should be counselled regarding the potential for local recurrence, there is no evidence that mastectomy improves their outcome.

Local recurrence is less common after BCS in older patients (>65 years). Recurrence is also less frequent in women with large breasts, but whether this relates to the larger excisions that can be performed in these patients or to alterations in steroid metabolism (fat is known to be an important site of conversion of androgens to oestrogens) is uncertain. A strong family history of breast cancer, and specifically carriage of a pathogenic mutation in one of the breast cancer genes, increases the risk of developing a second primary cancer in both the treated and contralateral breasts.

Tumour-related factors

Tumour location, tumour size, the presence of skin or nipple retraction and the presence or absence of axillary node involvement have not been shown consistently to predict for local recurrence after BCS.

Tumour phenotype

IBTR following BCS is more common in triple-negative and human epidermal growth factor receptor (HER2)-positive cancers. This has been confirmed by multiple studies, a systematic review and a meta-analysis. Regional recurrence and recurrence after mastectomy are also higher in patients with HER2-positive cancers and triple-negative breast cancers, and the type of surgery does not appear to influence outcomes ^, ( Fig. 7.2 ). A meta-analysis of 15 312 patients with triple-negative breast cancer showed fewer local recurrences (RR 0.75, 95% CI 0.65–0.87, P < 0.0001) and fewer distant metastases (RR 0.68, 95% CI 0.60–0.76, P < 0.00001) for BCT compared with mastectomy, demonstrating that tumour phenotype is not a reason to choose mastectomy over BCS, rather it is a reason to choose BCS rather than mastectomy.

Figure 7.2, (a) Cumulative local recurrence rate and (b) locoregional regional free recurrence rates after breast conserving surgery by tumour phenotype. 66 , 67 ER , oestrogen receptor; HER2 , human epidermal growth factor receptor 2.

Tumour size

Tumour size is not significantly associated with local recurrence. ^,

Tumour grade

A number of reports have analysed the relationship between tumour grade and local recurrence.

The lowest rates of local recurrence are reported in grade 1 tumours.

Although some series report a higher recurrence rate in grade 3 compared with grade 2 cancers, this is by no means universal. ^, ^, The RR of local recurrence between grade 1 and grade 2/3 cancer is approximately 1.5. The British Association of Surgical Oncology undertook a trial that randomised patients with node-negative grade 1 or special-type cancers to no further treatment, tamoxifen alone, radiotherapy alone or both radiotherapy and tamoxifen. An update of this study reported an exceedingly low rate of recurrence in patients randomised to radiotherapy and tamoxifen, and acceptably low rates of annual recurrence in patients treated with either tamoxifen alone or radiotherapy alone. Higher rates of recurrence were seen in patients who received neither radiotherapy nor tamoxifen. In these low-risk cancers, treatment by radiotherapy alone or tamoxifen alone can produce an acceptable rate of long-term control.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here